基于对象的合成孔径雷达影像极化分解方法

面向对象方法已成为全极化合成孔径雷达（SAR）影像处理的常用方法,但是极化分解仍以组成对象的像素为计算单元,针对以像素为单位的极化分解效率低的问题,提出一种面向对象的极化分解方法。通过散射相似性系数加权迭代,获得对象的极化表征矩阵并对其收敛性进行了分析,以对象极化表征矩阵的极化分解代替对象区域内所有像素的分解,提高极化特征获取效率。在此基础上,综合影像对象空间特征,并通过特征选择与支持向量机(SVM)分类进行分析和评价。通过AIRSAR Flevoland影像数据实验表明,面向对象的分解方法能够减少对象极化特征提取的时间,同时提高地物目标的分类精度。相对于监督Wishart方法,提出方法的总体精度和Kappa值分别提高了17%和20%。     

结合纹理与极化分解的面向对象极化SAR水体提取方法

合成孔径雷达(Synthetic Aperture Radar,SAR)拥有全天时全天候的工作能力,能够有效地连续对地观测,是土地管理、水体监测、灾害评估等多种应用的稳定数据来源。基于面向对象的思想,提出一种高精度、低虚警率的极化SAR(Polarimetric SAR,PolSAR)水体提取方法。此方法首先对极化SAR图像进行分割,再结合纹理与极化分解特征,对分割区域进行投票,识别水体区域。利用Radarsat-2数据和TerraSAR-X数据开展实验,并将提出方法与基于单一纹理和基于极化分解等水体提取方法进行对比,结果表明该方法在两种数据中均具有最高的总分类精度,其中基于分割技术能够保持完整的水陆边界,纹理与极化特征能够区分浅草、裸地和阴影等与水体相似的地物,结合投票方法能够提高小型水体检测率。     

基于商空间粒度的极化SAR图像分类

当前极化合成孔径雷达(SAR)图像的分类研究中,极化信息的不完全利用是影响极化SAR图像分类效果的重要原因之一。故将商空间粒度合成理论引入到极化SAR图像分类中,通过建立不同的支持向量机(SVM)分类器构建不同的商空间,从多个粒度层面实现对极化信息的综合利用。首先通过不同的极化分解方法得到不同的极化特征,分别对其建立不同的支持向量机分类器进行分类;再根据粒度合成理论对这些商空间进行融合,得到更细粒度上的改进的分类结果。最后,利用AIRSAR图像进行实验比较,算法改进后的结果在地物误分上有明显的抑制,各类别分类正确率都有所提高。     

Single‐look classification accuracy for polarimetric SAR

Polarimetric Synthetic Aperture Radar (SAR) systems such as ALOS‐PALSAR and Radarsat‐2 can operate in many different modes. The use of additional polarizations may require additional time and operating power and it is important to justify this by increased classification accuracy. A fully polarimetric, dual frequency AIRSAR scene from a rice‐growing area in Japan was classified by a maximum likelihood method based on the Wishart distribution. It is shown how the measured covariance matrices determine the separation accuracy between two classes. Closed form expressions are then given for the expected single‐look accuracy of the maximum likelihood classifier as a function of the class covariance matrices. This can be used to quickly compare the high spatial resolution classification performance of different polarimetric systems to decide upon a particular operating mode.     

基于多层支持向量机的极化合成孔径雷达特征分析与分类

为了充分利用极化合成孔径雷达（SAR）图像不同极化特征对不同地物目标类型的刻画能力,提出一种基于多层支持向量机（SVM）的极化SAR特征分析与分类方法。该方法首先通过特征分析确定适合不同地物类型的最佳特征子集;然后采用分层分类树的方式,根据每一种地物类型的特征子集逐层进行SVM分类;最终得到整体分类结果。RadarSAT-2极化SAR图像分类实验结果表明所提方法水域、耕地、林地、城区4类地物分类精度为85%左右,总体分类精度达到86%。该算法充分利用了不同地物目标类型的特性,提高了分类精度,也降低了算法时间复杂度。     

基于极化相似性特征的极化SAR图像的谱分类

针对极化SAR图像分类存在的问题,提出了基于SAR目标的极化特征的二维谱聚类方法。该方法可以充分考虑目标的极化相似性特征,利用二维的谱聚类方法实现极化SAR图像的分类。它以两目标散射的极化相似性参数图像作为输入特征,用二维图权函数代替一维图权函数求权值,使采样点分类和特征矢量分类相一致,从而实现极化SAR图像的分类。实验结果表明,该方法具有更好的分类结果,明显优于K均值分类。     

基于独立分量分析的极化SAR图像的相干斑抑制

研究基于独立分量分析( ICA)的极化合成孔径雷达（SAR）图像相干斑抑制方法。该方法将极化SAR图像斑点噪声的乘积模型,变换为应用ICA的信号独立加噪模型。并且将HV/VV的比值图像,也作为ICA的输入数据。利用ICA 的分离性,得到了分别对应于HH、HV和VV极化的三幅降噪图像。经本文方法处理后的图像,其相干斑噪声得到了有效的抑制,具有较高的等效视数,明显地改善了图像的质量。     

SAR图像船只分类识别研究进展

随着新一代SAR传感器的出现及应用,利用真实高分辨率、多极化SAR图像进行船只分类识别成为海上交通、渔业监测及国防应用的热点问题。首先回顾了SAR图像船只分类识别技术的发展。以近20 a国内外研究的重要成果为基础,对几何结构特征、散射特征和极化特征等船只分类特征进行了总结和比较,分析了其优缺点;总结了各种SAR图像船只分类识别算法的技术特点,并分析了各方法的适用性。最后对SAR图像船只分类识别技术的应用前景和发展趋势进行了阐述。     

A novel approach to polarimetric SAR data processing based on Nonlinear PCA

In remotely sensed Synthetic Aperture Radar (SAR) images, scattering from a target is often the result of a mixture of different mechanisms. For this reason, detection of targets and classification of SAR images may be very difficult and very different from other sensor imagery. Fully polarimetric data offer the possibility to separate the different mechanisms, interpret them and consequently identify the geometry of the targets. To achieve this task, several target decomposition techniques have been proposed in the literature to improve the interpretation of this kind of data. Among these, the physical based techniques are the most considered.     

基于多极化星载SAR数据的水稻/旱田识别-以江苏省海安县为例 

合成孔径雷达(SAR)数据对于南方多云多雨天气的地表农作物类型的探测具有独特的优势。以江苏省海安县为例,基于多极化SAR数据,包括双极化ALOS PALSAR以及全极化Radarsat\|2数据,采用面向对象的方法,针对当地水稻/旱田进行识别。针对双极化SAR数据,利用了其强度信息进行分类识别;而基于全极化数据,除强度信息外,还利用了其SAR信号统计分布概率进行分类规则建立。结果表明:L波段的ALOS PALSAR在识别旱地的桑树方面具有很大的优势,而基于两种分类方法的C波段Radarsat\|2数据识别水稻的精度分别为85%和75%,略低于ALOS PALSAR的识别结果(87.5%)。     

Land use classification with textural analysis and the aggregation technique using multi-temporal JERS-1 L-band SAR images

This paper describes a method of improving classification accuracy when using Synthetic Aperture Radar (SAR) images. The classifier used is a maximum likelihood classifier. Texture and textural feature images were made and used for classification. The accuracy of various classification methods was compared. As a result, it was found that the best classification was produced by the aggregation of the classified image when using texture images as additional inputs to the classifier. It is also shown that textural analysis and the aggregation technique are useful in the classification of SAR images.     

Classification of SAR images using morphological texture features

This paper presents applications of Synthetic Aperture Radar (SAR) image classification using morphological texture features. The texture features are based on morphological residues of opening and closing by reconstruction. It is shown that this set of features shows high 'robustness' to speckle perturbation in SAR images compared with those derived from traditionalmorphological residues. An algorithm based on estimating the divergence between and within classes was constructed in order to search for a discriminating feature subset. Higher classification accuracy was obtained by the optimized feature subset than by using other feature subsets derived from some well known texture characterization approaches. The classification accuracy was continuously improved by the introduction of post-processing filtering.     

Multi-chromatic analysis polarimetric interferometric synthetic aperture radar (MCA-PolInSAR) for urban classification

One of the problems of Synthetic Aperture Radar (SAR) polarimetric decomposition, is that oriented urban areas and vegetation signatures are decomposed into the same volume scattering mechanism. Such indetermination makes it difficult to distinguish vegetation from the oblique urban areas with respect to the radar illumination direction within the volume scattering mechanism. This event occurs because oriented targets exhibit similar polarimetric responses. This paper presents an improvement of the PolSAR decomposition scheme which permits the performing of more accurate classification. The method uses the information existing form the interference generated between two Doppler sub-aperture SAR images. This interferometric polarimetric SAR (PolInSAR) multi-chromatic analysis (MCA-PolInSAR) signal processing method permits the efficient separation of oriented buildings from vegetation yielding considerably improved results in which oriented urban areas are recognized, from volume scattering, as double-bounce objects. Results also show a considerable improvement in the robustness of classification and also in terms of definition and precision.     

滤波器组实现SAR图像中主要道路提取

合成孔径雷达(SAR)图像中道路目标自动提取既是当前遥感技术应用中的热点又是急待解决的难点.着眼SAR图像中道路目标的自动提取,在分析SAR图像特征的基础上,绕开常见的利用边缘检测算子提取图像中道路线段的作法,通过适当选择滤波窗口在去噪的同时尽量保持道路边缘,再通过二次二维方向性滤波、去枝滤波准确高效地提取了SAR图像中的道路目标.实验证明该方法在提取SAR图像中主要道路目标时几乎不需要调整参数,人工干预少,自动化程度高.     

基于Laplacian正则化最小二乘的半监督SAR目标识别

提出了一种基于核主成分分析(kernel principal component analysis,简称KPCA)和拉普拉斯正则化最小二乘(Laplacian regularized least squares,简称LapRLS)的合成孔径雷达(synthetic aperture radar,简称SAR)目标识别方法.KPCA特征提取方法不仅能够提取目标主要特征,而且有效地降低了特征维数.Laplacian正则化最小二乘分类是一种半监督学习方法,将训练集样本作为有标识样本,测试集样本作为无标识样本,在学习过程中将测试集样本包含进来以获得更高的识别率.在MSTAR实测SAR地面目标数据上进行实验,结果表明,该方法具有较高的识别率,并对目标角度间隔具有鲁棒性.与模板匹配法、支撑矢量机以及正则化最小二乘监督学习方法相比,具有更高的SAR目标识别正确率.此外,还通过实验分析了不同情况下有标识样本数目对目标识别性能的影响.     

全极化合成孔径雷达海浪遥感方法

提出了一种利用全极化真实孔径雷达测量方位方向和距离方向海浪斜率的新方法,该方法不同于极化方向角估计海浪方位向斜率方法,利用两种线极化图像信息将有关弱极化项剔除(流体动力学调制),获得仅包含倾斜项和极化调制项的雷达成像公式。利用机载SAR取代真实孔径雷达进行海浪反演,获得了与浮标致的结果。另外,在速度聚束模式情况下,推导了双极化合成孔径雷达图像谱同海浪谱的非线性变换关系。     

Fusion of multisensor multitemporal satellite data for land cover mapping

Efficient integration of remote sensing information with different temporal, spectral and spatial resolutions is important for accurate land cover mapping. A new temporal fusion classification (TFC) model is presented for land cover classification, based on statistical fusion of multitemporal satellite images. In the proposed model, the temporal dependence of multitemporal images is taken into account by estimating transition probabilities from the change pattern of a vegetation dynamics indicator (VDI). Extension of this model is applicable to Synthetic Aperture Radar (SAR) images and integration of multisensor multitemporal satellite images, concerning both temporal attributes and reliability of multiple data sources. The feasibility of the new method is verified using multitemporal Landsat Thematic Mapper (TM) and ERS SAR satellite images, and experimental results show improved performance over conventional methods.     

A comparison of fuzzy and neuro-fuzzy data fusion for flooded area mapping using SAR images

A comparison of change detection approaches for flooded area mapping using Synthetic Aperture Radar (SAR) images is provided. The aim was to assess the usefulness of fuzzy and neuro-fuzzy techniques for classification of SAR data. The work addresses both options of data-level fusion and decision-level fusion. The former is realized with multitemporal fuzzy or neural classification and the latter by combining classifications or fuzzy memberships for the pre- and post-event images. Highest overall accuracy values and flooded area accuracy values (90.3% producer's, 71.9% user's) were obtained from the neuro-fuzzy approach.     

基于谱归一化生成对抗网络的目标SAR图像仿真方法

为解决合成孔径雷达(Synthetic Aperture Radar, SAR)自动目标识别(Automatic Target Recognition, ATR)中的数据稀疏问题,提出一种基于谱归一化生成对抗网络(Spectral Normalization Generative Adversarial Network, SN-GAN)的目标SAR图像仿真方法。本文方法通过构建目标—场景—雷达耦合物理模型,求解散射强度分布图,利用SN-GAN实现对散射强度分布图的优化,生成高质量仿真SAR图像。通过3种相似性评估算法对仿真图像进行相似度评估,验证本文仿真方法的有效性。最后通过多组SAR ATR进行实验验证,在训练集中加入SN-GAN优化的仿真SAR图像可以有效缓解数据稀疏问题,提升分类算法的准确率。     

极化SAR海面船只检测技术的研究进展

极化SAR技术是近年SAR领域的研究热点。首先介绍了不同极化方式对船只、海洋杂波以及船只尾迹在SAR图像上的成像影响,详细概括和总结了目前应用于海洋船只监测的各种极化SAR技术特点,最后分析了极化SAR海洋船只监测技术的应用前景。